Why the simulation stops when the agent is executing?

[NTMPhg13]

Thanks for your amazing work with this repo and paper. I'm having difficulty understanding SUMO. I notice when the agent is executing with the tools , the vehicles on SUMO kind of stop moving, they just stay still until the decision of next phase is decided. That looks like SUMO is lagging. Do you know if there are any problems with this or is that normal?. Thanks

[pangay]

Hi, this is actually normal behavior. SUMO will not proceed to the next simulation step until it receives the control command from your agent. Since the decision-making process (especially when involving large models) may take a relatively long time, the vehicles appear to “freeze” during that period.

So the pause you observed is not SUMO lagging, but simply waiting for the next action from the controller.